Before looking in more detail at the patterns of state-sponsored violence during the period in which Siyad Barre’s held power (October 1969 to January 1991), I would like to make some general comments about political violence an other useful concepts in the Somali context.

A) Some Preliminary Notes on Violence, Clan and Politics in Somalia
1- Although this seminar is concerned with political violence, we need to bear in mind that nothing political takes place in a social vacuum and that violence with ostensible political goals is embedded in a broader violent context that comes to symbolize an epoch.

As a matter of fact, the collapse of what remained of political institutions in January 1991 resulted in all kinds of violence developing alongside proper political conflict including criminal violence, sexual violence (in particular against women), personal revenge and pathological behaviors of all sorts. Therefore, not all violence was politically motivated, but it was sometimes difficult to decipher whether an incident (for example car hijacking or kidnapping of foreign workers) was political or not. The political context itself stimulated criminal activities.

Yet, given the clan structure of the Somali society, then the last effective ideological and practical frame of reference that remained for most people once the state and all modern institutions had collapsed, it was difficult to see any such violence as “private”—i.e. disconnected from clan-based factional politics. The picture was so blurred that the violence of clan conflicts has been overemphasized—thus pointing at the “exceptionality” of the Somali case.

2- We need to distance ourselves from the emotional reactions to Somalia’s civil war that prevailed in the international press and many NGO reports in the 1990s. We must put violence in a comparative perspective, both historically and culturally. There had been instances of cruel clan warfare in the past, before and during colonial time (let’s recall Sayid Mahamed Abdulle Hassan’s use of mass violence against clan segments collaborating with colonial power or perceived as hostile),[1] as much as there were institutions such as councils of elders and diya payments to settle scores. It was increasingly popular in the 1990s to idealize in retrospect Somali society prior 1969 seen as inherently pacific, despite significant evidence to the contrary in colonial administrative reports, travelers’ accounts and anthropological works.

It is also too convenient to forget the amount of violence used by colonial powers to “pacify” different areas of the Horn, in particular by the Italians during the fascist regime. Mahamed Siyad Barre was socialized during that era, in the Italian colonial police, and he probably took part in the invasion of Ethiopia in which 40,000 Somalis troopers were involved. He knew very well what could be gained from the unrestrained use of violence by the state. The Soviet patronage the Somali enjoyed in the 1970s did nothing to modify his views.

When we look at the extent of mass atrocities and ethnic cleansing that took place in Europe, in the wake of Word War II,[2] or more recently in former Yugoslavia, let alone in the current Syrian crisis, the Somalis’ political violence does not look so extreme. What is more unusual is the length of the crisis in southern Somalia.

3- Although there is some causal link between state collapse and the level of violence that engulfed the country after the fall of Villa Somalia and the flight of Siyad Barre, we should not oversimplify the correlation between these two variables. Curbing political violence was certainly an important precondition for restoring a form of government in autonomous Somaliland. However, violence sometimes leads to the establishment of stable forms of state power (see the success of EPRDF that seized power also in 1991 in Ethiopia). A political entrepreneur using violence to establish his domination should be able to alter the balance of power in a meaningful and durable way, something no clan-based faction or coalition of factions was ever able to achieve in southern Somalia after January 1991. Foreign intervention interfered with this process in a counterproductive way, without providing effective alternatives. Addressing the issue of violence separately, without taking into account the political process, might bias our understanding of the situation and shift the debate towards moral issues.

4- In understanding the roots of political violence, and the vicious circle of its repetition, we should avoid naturalizing Somalis’ identity (today as war addicts), as it was done in the past when they were once portrayed as “pastoral democrats”, “fierce nationalists”, “moderate Muslims” and so on. Well-wishing observers, including foreign academics, projected their own prejudices and expectations on the Somali complex social system,[3] and these constructs were deeply misleading. The ability of Somali politicians to rise above clan politics in the 1960s and 1970s was overestimated (except by IM Lewis who, however, contradicted himself by supporting the idea of a Somali nation united by language and custom—a political project rather than a fact to this date).[4] The Somalis’ problematic relationship to the imported nation-state framework forced upon them by Western colonization was never properly weighed, although it is a trivial enough observation for many segmentary societies in social anthropology.

These constructs still inform the dominant discourse about the restoration of a “national government” in Somalia, and form the background of the doomed (and largely bloody) attempts at political engineering that pervaded the past two decades. Understanding “patterns of violence” in contemporary Somalia requires departing from teleological perspectives about state building as a prerequisite for peace.

5- Siyad Barre’s regime could not be characterized as the domination of one clan–or even the so-called MOD[5] alliance–over the rest of the society. Had it been the case, it would have disintegrated far sooner because clan alliances are usually instable. As I argued at length in my PhD thesis, it was apersonal rule relying on an extended client system of patrimonial servants cemented by fear, greed and marriage. Indeed, the security apparatus became increasingly controlled by officers from these three clans, and at the end of the period, Siyad Barre’s hold of power depended on a few well-armed units commanded by Marehan officers—mainly from his Rer Koshin Dini. However, both the military and civilian state elite included individuals from most clan segments, and many Hawiye, Dir, Majerten and Isaq businessmen got very rich under Siyad Barre. These people did not “represent” their clan in government as the elected politicians did to a certain extent in the 1960s. They were used nevertheless by Siyad to isolate those targeted by state repression at a given time, and provided a channel to send messages and gifts to the clan elders. Whatever the political reality, what counted eventually was the popular misperception. It explains why USC Hawiye militias hunted down all Darod people during the fighting of January 1991,[6] prompting the latter in the South to rally behind the former dictator. Indeed, Siyad Barre did not invent the mobilization of clan affiliations in political conflict. It was already a prominent feature of Somali politics before 1969. However, the extreme politicization of clan segmentation by his regime has transformed the nature of clan warfare in Somalia, adding on the top of the existing layer of past feuding cycles—then mitigated by truces and intermarriage—a much deeper antagonism and an appetite for revenge for the crimes committed during 21 years. This is one of Siyad Barre’s lasting legacies: his politics instilled distrust and hatred even among long time friends and family members.

B) Patterns of State Violence under Siyad Barre
In spite of the fairy tales about the “progressive” era of the 1970s, violence or the threat to use it was consubstantial to the regime, although it was less visible from 1969 (“the bloodless revolution”) to 1978. It is difficult to evaluate the extent of genuine support the regime then enjoyed, when a full dictatorship solidified as early as 1971 with the backing the Soviet Union, a powerful patron with its ubiquitous advisers in the state bureaucracy and the parastatals. A major exception to the lack of visible opposition was the protest movement in the mosques against the 1974 law on women’s status and the family perceived as undermining the Islamic law (e.g. banning polygamy and giving equal rights to women in inheritance). The spontaneous movement was suppressed heavy-handedly: many imams and sheikhs were detained and tortured, and ten alleged leaders were publicly executed in January 1975. In June 1975 a major purge of the civil service and the army—the second since 1969—targeted people who had disagreed publicly (or were suspected of having done so) with these murders.

Throughout the 1970s state violence took the form of harassment and arbitrary detention of thousands of people—far beyond the circle of rivals and true opponents—by the fast growing surveillance apparatus making good use of its training in the Soviet block countries. Prominent were the ubiquitous political police (the NSS) backed by the extra-judiciary National Security Court, the military police (HANGASH) created in 1978, which provided the presidential guard and played an increasing role in the civil war of the late 1980s, and the Youth militia called Gulwadayal. Uniformed police and army units could also be involved. Various observers, including academics, largely underestimated the extent of this police state at the time. It generated an atmosphere of fear and distress, and legitimized the opposition’s later use of violence.

Unlike the 1964 border war, the full-fledged invasion of Ethiopia in 1977 that led to a resounding defeat for the Somali army had a significant impact on society. The core of the regime’s legitimization was the nationalist discourse on Greater Somalia, and many perceived the Soviet alliance as a temporary nuisance necessary to achieve this objective. The shattered dream—for a foreseeable future—and the loss of Soviet patronage eroded Siyad Barre’s legitimacy. Moreover, he was widely held responsible for both political and strategic mistakes that left Somalia weakened and humiliated, saved from a shameful Ethiopian occupation by a last minute deal between Washington and Moscow. The widespread unrest in the army led to the April 9, 1978 coup attempt. Although its leader, Abdullahi Yusuf Ahmed was Umar Mahamud, the core group came from various clans including Hawiye and Isaq. However the state repression focused on the Majerten,[7] because from the onset, Siyad Barre had been wary of this clan’s potential threat. He had antagonized the Majerten while working for the Italians in the 1950s, when many radical nationalists came from that clan. The SYL candidate to succeed President Shermarke, assassinated on October 15, 1969, Haji Muse Boqor, belonged to the same sub-clan and many Majerten felt that the top job was stolen from them. Although the original military junta included two Majerten members, Siyad made sure to leave out of the SRC the most prominent colonels belonging to that clan.

When Abdullahi Yusuf created the SSF in February 1979 and then the SSDF in October 1981, and launched a guerrilla force from the Ethiopian side of the border, Siyad’s security forces assaulted the Umar Mahamud living in Mudug and Nugal regions, killing people, raping women, destroying settlements, slaughtering livestock and poisoning the wells. This was the first “war against his people” launched by Siyad Barre, years before the near annihilation of the Isaq in the North, and the first time in modern Somali history when the state intervened in the bush not to quell inter-clan warfare but to fuel it. Like a self-fulfilling prophecy, targeting the Majerten seemed to give credit to the government propaganda of a coup motivated by “tribalism”, which helped dividing the growing opposition.[8] Violence against civilians was meant also to dissuade the Majerten from crossing the border in mass to join the SSDF guerilla.

Isaq officers were active in the new regime in the 1970s, and played a significant role in the Ogaden war. Isaq businessmen who had invested in Mogadishu in the late 1960s were prominent in the import/export trade. However, Siyad Barre treated the clan family as an enemy as soon as the SNM was created in April 1981 and vowed to overthrow the regime. At about the same time, some civilians, who had organized a true self-help movement in Hargeysa (the Uffo group) to protest against the state neglect of the North, were arrested. Their trial in February 1982—with confessions extracted by torture—generated popular unrest especially among the youth, with dozens killed and hundreds detained and severely beaten. The students kept organizing demonstrations in the towns in 1983 and 1984 with a more political agenda leading to more repression. Then some Isaq elders sponsored the army deserters who founded the SNM guerilla force in Ethiopia.

However, it is the disproportionate collective punishment carried out by the regime that generated the support for the liberation front: after every guerilla attack the NSS and the army retaliated with arbitrary killings, arrests and tortures, behaving as a foreign force occupying the country. Harassment including rape, beatings, racketing, and the looting of Isaq properties became the norm. When, Siyad Barre’s peace agreement with Ethiopia in April 1988 forced the SNM guerrillas to cross the border into Somaliland for a desperate offensive in May, the government forces once again overreacted to break the popular uprising by bombing Hargeisa and Burao, indiscriminately killing civilians (15 to 20,000), including columns of displaced people fleeing the combat zone. Although the outgunned SNM had to evacuate Hargeysa and Burao in mid-August 1988, the civilians had fled the ghost towns and the regime never regained full control of the North.

In the early 1980s Siyad ordered the arming of some Dulbahante sub-clans and Ogadeni refugees in the northern camps[9] to fight the Habar Yunis and Idagalle, and then a Gadabursi militia to fight the Habar Awal. When the SSDF fizzled out in 1985 and Abdullahi Yusuf was detained in Ethiopia (until 1991), defectors from this front and other Majerten were drafted into the army to fight the SNM, especially after general Mahamed Said Hirsi “Morgan” was appointed head of the military region that encompassed the Isaq territory in Somaliland. Such policy of arming clan militias was extended to the southern part of the country when the Ogadeni SPM in Jubaland and the Hawiye USC in Hiran opened new guerilla battle lines.[10] Unpaid soldiers operating in the North survived through looting and ransoming the local people, sometimes selling their weapons to their opponents in the liberation fronts. In this context of anarchy, they were mass desertions of Isaq, Hawiye and Ogadeni soldiers from mid 1988 onward to join their respective guerilla organizations. The regime retaliated with massacres of civilians from the same lineages. The disintegration of the army largely explains why the SNM was able to take control of the North in late 1990 and early 1991, and why the USC guerilla force managed to fight its way to Mogadishu so fast from June 1990 to January 1991. This disintegration is a by-product of Siyad Barre’s use of unrestrained, clan-targeting violence.

Hasty recruitment of MOD and increasingly Marehan lineages into the army transformed this formerly professional corps—still able by 1982 to contain the joint SSDF and Ethiopian invasion—into an aggregate of clan militias by the late 1980s, which were proficient in harassing and killing civilians only. Their lack of professionalism and discipline, and prevailing clan loyalties led to an increasing amount of war crimes and criminal behaviors. For example, the presidential guard recruiting young Marehan from the bush became reckless and committed atrocities and murders in Mogadishu—such as the assassination of the Catholic archbishop in July 1989—with total impunity. Civil protest in Mogadishu, such as the Islamic-led demonstrations in mid-July1989, was handled brutally with heavy fire at the demonstrators (several hundred casualties in one day), mass detention and torture. The more threatened the regime felt the harsher the forms of repression. However, the growing abuses by the presidential guard precipitated the Hawiye uprising in the capital city in December 1990.

We should not underestimate the impact on the moral fabric of the society of the dissemination of weapons, the inter-clan fighting and the repeated, often sadistic atrocities that characterized the later years of Siyad’s regime. The perpetrators of violence were never held responsible of their deeds. It became legitimate to kill for a political purpose or to avenge your kin. Banditry developed in both urban and rural areas before the fall of Siyad. Human life had little value. Young adults who grew under Siyad’s rule equated state power with treachery and violence. Indeed, the atrocities perpetrated by the clan militias in the 1990s mimicked the recurrent behavior of the security apparatus since the late 1970s. The collapse of the schooling system, the ruined economy and the empty state coffers led many young men to use their weapons acquired for self-protection to obtain what they needed—announcing the moryans of the 1990s.[11] Looting became the most common form of salary for combatants.

Because Siyad refused to leave power when given the option or to make sincere concessions—to the Manifesto group for example—until he was forced to flee by the USC uprising, and because he destroyed the state as well as the economy through his strategy of political survival, the subsequent two decades of anarchy are as much a product of Siyad’s regime as the consequence of the liberation front’s irresponsible behavior. Unlike in Ethiopia, where the EPRDF was able to use the predominantly Amharic state apparatus to tighten its grip, there were no government institutions left in Somalia over which the victors could take control.

To promote an era of peace and reconciliation in Somalia, facilitate a negotiated political settlement based on justice, and to end the culture of impunity, it is important to acknowledge this history and its enduring legacy.


Daniel Compagnon is Professor of Political Science at Science Po Bordeaux


[1] The ‘Mad Mullah’ was defeated in 1920 only, when the colonial air force bombed his fortified stronghold. The jihad and British counterinsurgency operations caused an estimated 200,000 casualties.

[2] See Keith Lowe’s recent synthesis, Savage Continent: Europe in the Aftermath of World War II, Penguin Books 2013 (Viking 2012).

[3] That some brilliant, Western-educated Somali intellectuals were the strongest supporters of such myths did nothing to dispel the fantasy until well into the 1990s. Many of these intellectuals later became unashamed clan chauvinists.

[4] Even the linguistic and cultural divide between pastoral nomads from the four, Samale clan families, and semi-sedentary Sab agriculturists living in the South, let alone the peasant communities of Bantu origin, was deliberately downplayed.

[5] For Marehan, Ogaden, Dulbahante, three clans of the Darod clan family.

[6] Ali Mahdi and other Hawiye “moderates” did nothing to prevent the assassination of Darod members of the Manifesto group by Hawiye militias. The disintegration of the Manifesto along clan affiliations and USC-Aydid’s attack on the SPM triggered the subsequent clan-based factionalism.

[7] Among the hundreds of military arrested, 17 officers predominantly Majerten were selected to be executed in July1978.

[8] A decade later many Hawiye still used it to justify their inaction and rejection of the SSDF.

[9] Siyad’s regime’s abuse of the HCR resources and the drafting of refugees into pro-government militias was documented at the time by various reports but it remains under-researched. The SNM attacked the Ogaden refugees when taking control of Somaliland, both those remaining in the camps and those “resettled” in the towns.

[10] During its advance towards Mogadiscio, the USC attacked the refugee camps around Beled Weyne and Jalalaqsi, from which a pro-government militia assaulted their columns.

[11] See Roland Marchal’s work on the moryans of Mogadishu.


Tagged with:

One Response to State-sponsored violence and conflict under Mahamed Siyad Barre: the emergence of path dependent patterns of violence

  1. […] Compagnon unpacks clan violence in Somalia and how Siad Barre used clan violence as a means of […]

Leave a Reply

Your email address will not be published.